Fault prediction method and device of server, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a server fault prediction method and device, a storage medium and electronic device.The method comprises the steps that various kinds of operation information generated by a server in the current service handling process are obtained, each kind of operation information corresponds to a server component deployed in the server, and the server component is configured to execute the operation of the server in the current service handling process; each type of operation information is used for indicating the operation state of the corresponding server component in the current service handling process; target service information of each server component is predicted according to the operation information, and the target service information is used for indicating the influence condition of the current operation state of the server component on the service handling quality of the server; and predicting target fault information of the server according to the target service information of the plurality of server components. Through the server fault monitoring method and device, the problem that the fault monitoring efficiency of the server is low in the related technology is solved, and the effect of improving the fault monitoring efficiency of the server is achieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and in particular, to a method, device, storage medium, and electronic device for predicting server failures. Background Art

[0002] With the rapid development of big data and artificial intelligence technologies, servers, as core devices of modern information systems, undertake the important task of storing and processing large amounts of data. However, due to the high-load operation of servers and the interference of various external factors, server failure has become an issue that cannot be ignored. Therefore, the quality of server business processing can be guaranteed by monitoring the operation of the server to avoid the impact on the business handled on the server due to the failure of the server to be perceived. In the related art, the way to monitor server faults is usually to deploy a health light on the server. When the server fails, the health light will light up. Although this method can indicate faults, it is an indication after the server has failed. That is, although the fault has been monitored, the business handled on the server is also affected at this time, and the maintenance personnel have to troubleshoot after receiving the indication, resulting in low efficiency of server fault monitoring.

[0003] With regard to the problem of low efficiency of server fault monitoring in related technologies, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method, device, storage medium and electronic device for predicting server faults, so as to at least solve the problem of low efficiency of server fault monitoring in the related art.

[0005] According to an embodiment of the present application, a method for predicting a server failure is provided, comprising:

[0006] Acquire a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process;

[0007] Predicting target service information of each of the server components according to the operation information, wherein the target service information is used to indicate the influence of the current operation status of the server component on the quality of server service handling;

[0008] The target fault information of the server is predicted based on the target business information of the plurality of server components, wherein the target fault information is used to indicate an operation fault that will occur in the server during the current business handling process.

[0009] Optionally, predicting target service information of each server component according to the operation information includes:

[0010] Generating an initial component score of the server component according to the operation information, wherein the initial component score is used to indicate the impact of the operation state indicated by the operation information on the current service handling quality of the server;

[0011] The initial component score is weightedly calculated using the initial weight parameters corresponding to the server component to obtain a target component score of the server component, wherein the target business information includes the target component score, and the initial weight parameters are used to indicate the influence of the component function of the server component on the business handling quality of the server.

[0012] Optionally, generating the initial component score of the server component according to the operation information includes one of the following:

[0013] In the case where the operation information is the first information, the first information is matched with a preset information range to obtain a target information range within which the first information falls, wherein the first information is the operation data reported by the server component when it is determined that the server component is currently in an abnormal operation state; and the component score configured corresponding to the target information range is determined as the initial component score;

[0014] In the case where the operation information is the second information, an operation information change curve of the server component before the current moment is generated according to the second information, wherein the second information is the operation data of the server component collected before the current moment, and the operation information change curve is used to indicate the change of the operation data of the server component in a continuous time period before the current moment; the operation information change curve is matched with a preset information change curve to obtain a target information change curve whose matching degree with the operation information change curve is greater than or equal to a target matching degree; and a component score configured corresponding to the target information change curve is determined as the initial component score;

[0015] In the case where the operation information is third information, the initial component score corresponding to the third information is determined from information and component scores having a corresponding relationship, wherein the third information is used to indicate the number of times the server component is in an abnormal operation state during the server handling of business.

[0016] Optionally, predicting target fault information of the server according to the target service information of the plurality of server components includes:

[0017] Filtering a target component from the server components according to the target business information, wherein the target business information corresponding to the target component is used to indicate that the degree of influence of the current running state of the target component on the service handling quality of the server is greater than the target influence degree;

[0018] The target fault information corresponding to the target business information and the target component is screened out from the business information, component and fault information having corresponding relationships.

[0019] Optionally, after predicting target fault information of the server according to the target service information of the plurality of server components, the method further comprises:

[0020] Generate fault maintenance information for the server according to the target service information, wherein the fault maintenance information is used to prevent the server from having a fault indicated by the target fault information during the current service handling process;

[0021] Perform fault maintenance on the server according to the fault maintenance method indicated by the fault maintenance information.

[0022] Optionally, the generating fault maintenance information for the server according to the target service information includes:

[0023] Filtering a reference component from the server according to a target fault type to which the fault indicated by the target service information belongs, wherein the reference component is a server component associated with the operation fault indicated by the target fault information;

[0024] The fault maintenance information is generated according to the current operation information of the reference component.

[0025] Optionally, generating the fault maintenance information according to the current operation information of the reference component includes one of the following:

[0026] In the case where the target fault type is a processor occupancy fault, generating an occupancy adjustment operation for the processor according to the initial occupancy of each processor in the plurality of processors deployed in the server, wherein the reference component includes the plurality of processors, the operation information includes the initial occupancy, the fault maintenance information includes the occupancy adjustment operation, and the occupancy adjustment operation is used to adjust the difference between the occupancies of the plurality of processors by adjusting the occupancy of each processor;

[0027] In the case where the target fault type is server operation overtemperature, generating an operation parameter adjustment operation for the heat dissipation device deployed in the server according to an initial heat dissipation of the heat dissipation device, wherein the reference component includes the heat dissipation device, the operation information includes the initial heat dissipation, the fault maintenance information includes the operation parameter adjustment operation, and the operation parameter adjustment operation is used to adjust the heat dissipation of the heat dissipation device;

[0028] In the case where the target fault type is a target memory operation fault, the resource occupancy of a reference memory deployed in the server is obtained to generate a data transfer operation for the target memory, wherein the reference memory is a memory other than the target memory among multiple memories deployed in the server, the reference component includes the reference memory, the operation information includes the resource occupancy, the fault maintenance information includes the data transfer operation, and the data transfer operation is used to isolate the target memory and store the data stored in the target memory into the reference memory.

[0029] According to another embodiment of the present application, a server fault prediction device is provided, comprising:

[0030] An acquisition module, used to acquire a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process;

[0031] A first prediction module, configured to predict target service information of each of the server components according to the operation information, wherein the target service information is used to indicate the influence of the current operation state of the server component on the quality of server service handling;

[0032] The second prediction module is used to predict the target fault information of the server according to the target business information of the plurality of server components, wherein the target fault information is used to indicate the operation fault that will occur in the server during the current business handling process.

[0033] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0034] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0035] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0036] Through this application, by obtaining a variety of operating information generated by the server during the current business handling process, since the operating information can reflect the operating status of each component in the server during the current business handling process, it is possible to predict the target business information indicating the impact of each server component on the server business handling quality under the current operating state according to the acquired operating information of each server component, that is, the impact of the current operating state of each server component on the server business handling quality, and then, based on the predicted target business information of multiple server components, further predict the target fault information that characterizes the operating fault that will occur in the server during the current business handling process. Through the above method, the potential failure risk of the server can be predicted in advance according to the operating status of each server component in the server, and then corresponding prevention and treatment measures can be taken in advance according to the predicted fault information, avoiding the passive perception of the impact on the server business after the server failure occurs. Therefore, the problem of low efficiency of server fault monitoring in the related technology can be solved, and the effect of improving the efficiency of server fault monitoring can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a hardware structure block diagram of a server device according to a method for predicting server failures in an embodiment of the present application;

[0038] Figure 2 is a flow chart of a method for predicting server failures according to an embodiment of the present application;

[0039] Figure 3 It is a design diagram of a method for evaluating server health score according to an embodiment of the present application;

[0040] Figure 4 is a schematic diagram of a method for collecting health indicators according to an embodiment of the present application;

[0041] Figure 5 is a schematic diagram of scoring based on a weight matrix according to an embodiment of the present application;

[0042] Figure 6 is a workflow diagram of a health score arbitrator according to an embodiment of the present application;

[0043] Figure 7 is a schematic diagram of a hardware link topology according to an embodiment of the present application;

[0044] Figure 8It is a structural block diagram of a server fault prediction device according to an embodiment of the present application. DETAILED DESCRIPTION

[0045] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0047] The method embodiments provided in the embodiments of the present application can be executed in a server device or a similar computing device. Taking running on a server device as an example, Figure 1 1 is a hardware structure block diagram of a server device according to a server fault prediction method of an embodiment of the present application. Figure 1 As shown, the server device may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the server device may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above server device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0048] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the fault prediction method of the server in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the server device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0049] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the server device. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0050] In this embodiment, a method for predicting server failure is provided. Figure 2 is a flow chart of a method for predicting server failures according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0051] Step S202, obtaining a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process;

[0052] Step S204, predicting target service information of each of the server components according to the operation information, wherein the target service information is used to indicate the influence of the current operation status of the server component on the quality of server service handling;

[0053] Step S206: predicting target fault information of the server according to the target business information of the plurality of server components, wherein the target fault information is used to indicate an operational fault that will occur in the server during the current business processing.

[0054] Through the above steps, by obtaining a variety of operating information generated by the server during the current business handling process, since the operating information can reflect the operating status of each component in the server during the current business handling process, it is possible to predict the target business information indicating the impact of each server component on the quality of server business handling under the current operating state according to the acquired operating information of each server component, that is, the impact of the current operating state of each server component on the quality of server business handling, and then, based on the predicted target business information of multiple server components, further predict the target fault information that characterizes the operating fault that will occur in the server during the current business handling process. Through the above method, the potential failure risk of the server can be predicted in advance according to the operating status of each server component in the server, and then corresponding prevention and treatment measures can be taken in advance according to the predicted fault information, so as to avoid the impact of passive perception on the server business after the server failure occurs. Therefore, the problem of low efficiency of server fault monitoring in the related technology can be solved, and the effect of improving the efficiency of server fault monitoring can be achieved.

[0055] In the technical solution provided in the above step S202, the various operating information are various parameters used to characterize the operating status of the corresponding server components in the current business processing process. For example, if the server component is a CPU (Central Processing Unit), the operating information may include but is not limited to the number of threads, temperature, load, response speed, etc.; if the server component is a fan, the operating information may include but is not limited to the fan speed, air volume, noise level, etc., and this solution does not limit this.

[0056] Optionally, in an embodiment of the present application, obtaining various operating information generated by the server during the current business processing process can be actively collecting the operating information of each server component during the business processing process, or it can be receiving abnormal operating information actively reported by the server component when an abnormality occurs in the server component. This solution does not limit this.

[0057] Optionally, in an embodiment of the present application, in order to avoid excessive occupation of system resources and affecting business processing efficiency, when the operation information of the server components is actively collected, a fixed interval can be set to periodically collect the operation information, or whether to collect the operation information can be selected based on the performance status of the server during business processing. For example, before collecting the operation information of the server components, a health assessment of the performance status of the server during the current business processing can be performed to obtain the performance status data of the server. The performance status data may include but is not limited to server CPU usage, network delay and throughput, queue length, task response time, etc. Based on the above performance status assessment, specific collection trigger conditions are set. For example, when the average network delay of the server exceeds 100ms or the throughput is lower than a preset threshold, it is considered that the server network status is not good at this time, and the collection of the operation information of the server components is automatically triggered.

[0058] In the technical solution provided in the above step S204, the target business information is information used to characterize the degree of influence of the current operating status of the server component on the quality of server business processing. The target business information can be in the form of a score to reflect the degree of influence of the current operating status on the quality of server business processing. For example, the lower the score, the greater the impact. This solution does not limit this.

[0059] Optionally, in an embodiment of the present application, the predicted target business information may be generated by generating initial component scores for corresponding server components based on operating information, and then using weight parameters corresponding to the server components to perform weighted calculations on the initial component scores to obtain a comprehensive score. The comprehensive score can reflect the degree of influence of the current operating status of the server components on the quality of server business processing.

[0060] In the technical solution provided in the above step S206, the target fault information is used to reflect the operational failures that may occur in the server during the current business processing. The target fault information may include but is not limited to the type of faulty component, fault severity, fault cause, fault occurrence time, etc. This solution does not limit this.

[0061] As an optional embodiment, predicting target service information of each server component according to the operation information includes:

[0062] Generating an initial component score of the server component according to the operation information, wherein the initial component score is used to indicate the impact of the operation state indicated by the operation information on the current service handling quality of the server;

[0063] The initial component score is weightedly calculated using the initial weight parameters corresponding to the server component to obtain a target component score of the server component, wherein the target business information includes the target component score, and the initial weight parameters are used to indicate the influence of the component function of the server component on the business handling quality of the server.

[0064] Optionally, in the embodiment of the present application, an initial component score is first generated for the corresponding server component according to the operation information, and then the initial weight parameter corresponding to the operation status of the server component is used to perform weighted calculation on the initial component score to obtain the target component score. For example, taking the server component PSU (Power Supply Unit) as an example, when the operation information indicates that the current voltage of the PSU is too high, the PSU is assigned an initial component score of 80, and then the initial weight parameter corresponding to "PSU overvoltage" is 40% in the weight parameter table. At this time, it can be calculated that the final target component score of the PSU is 80*40%=32.

[0065] Optionally, in an embodiment of the present application, the initial weight parameter can be manually set based on experience, or the initial weight parameter can be dynamically adjusted based on the difference between the predicted fault information and the fault information actually detected by the maintenance personnel, which can be: collecting the actual fault information of the server from the server operation log, analyzing the difference between the actual fault information and the target fault information, adjusting the initial weight parameter according to the difference to obtain the target weight parameter, and then applying the target weight parameter to the next round of server fault prediction operations. For example, if it is found that the actual impact of failures caused by frequent CPU overheating is more serious than predicted, the CPU weight parameter can be increased, and if the false alarm rate of the disk fault prediction is high, the disk weight parameter can be reduced, or more disk health-related features can be introduced into the evaluation features to improve the prediction accuracy.

[0066] Through the above method, the weight parameters of server components can be dynamically adjusted based on the comparison between the actual fault situation and the predicted results, thereby improving the accuracy and reliability of server health status assessment and fault prediction.

[0067] As an optional embodiment, the generating the initial component score of the server component according to the operation information includes one of the following:

[0068] In the case where the operation information is the first information, the first information is matched with a preset information range to obtain a target information range within which the first information falls, wherein the first information is the operation data reported by the server component when it is determined that the server component is currently in an abnormal operation state; and the component score configured corresponding to the target information range is determined as the initial component score;

[0069] In the case where the operation information is the second information, an operation information change curve of the server component before the current moment is generated according to the second information, wherein the second information is the operation data of the server component collected before the current moment, and the operation information change curve is used to indicate the change of the operation data of the server component in a continuous time period before the current moment; the operation information change curve is matched with a preset information change curve to obtain a target information change curve whose matching degree with the operation information change curve is greater than or equal to a target matching degree; and a component score configured corresponding to the target information change curve is determined as the initial component score;

[0070] In the case where the operation information is third information, the initial component score corresponding to the third information is determined from information and component scores having a corresponding relationship, wherein the third information is used to indicate the number of times the server component is in an abnormal operation state during the server handling of business.

[0071] Optionally, in an embodiment of the present application, the first information is the operating data reported by the server component when it detects that it is in an abnormal operating state. The preset information range may be a numerical interval outside the normal operating range defined according to empirical values. For example, for the CPU temperature state, the preset information range may include different intervals such as "70-80°C", "80-90°C", and "90-100°C". After obtaining the first information, the first information is matched with the preset information range to find the target information range in which the first information falls. Each target information range is pre-configured with a corresponding component score, which reflects the degree of influence of the abnormal data on the quality of business handling. For example, regarding the temperature state of the CPU, the preset information range includes "70-80°C", "80-90°C", and "90-100°C". Since the higher the CPU temperature, the greater the degree of influence on the quality of business handling, the component scores corresponding to these three information ranges may be 80, 70, and 60, respectively, and the component score corresponding to the target information range in which the temperature information currently reported by the CPU falls can be determined as the initial component score.

[0072] Optionally, in an embodiment of the present application, the second information is the operation data of the server component that changes over time, such as disk read / write rate, fan speed, sensor reading, etc. The system can generate an operation information change curve of the server component in a continuous time period before the current time point. For example, a disk read / write rate change curve in the past 24 hours, a week, or a month, or a memory usage fluctuation curve in the past week can be generated. The change curve can be a time series graph that intuitively displays the change trend of data over time. The preset information change curve can be a pre-established curve library containing typical change curve patterns under various normal and abnormal operating conditions. For example, for the disk I / O rate, different curve patterns such as "stable read / write", "burst read / write", and "I / O bottleneck" may be preset. The generated operation information change curve is then matched with the preset information change curve library to find a curve with a matching degree greater than or equal to the target matching degree. The target matching degree can be a predefined threshold, which indicates the similarity between the change curves. The matching degree can be quantified by calculating the distance between the two curves (such as Euclidean distance, dynamic time warping DTW, etc.). The smaller the distance, the higher the matching degree. After obtaining the target information change curve with a matching degree greater than or equal to the target matching degree, the initial component score is determined according to the component score of the corresponding configuration of the curve.

[0073] Optionally, in the embodiment of the present application, the third information is the number of times the server component is in an abnormal operation state during the server processing business, such as the number of times the CPLD has an I ERR (uncorrectable error), the number of times the PSU has voltage instability, etc. A component score mapping table can be predefined according to the degree of influence of the number of abnormal operations and the quality of business processing. For example, if the CPLD has three I ERRs, the component score is set to 70 points, and if the CPLD has five I ERRs, the component score is set to 60 points. Then, the number of abnormal operations of the server component can be directly converted into the initial component score through the above mapping table.

[0074] Through the above method, which combines multiple methods such as abnormal data monitoring, operation trend analysis and abnormal frequency statistics, it is possible to comprehensively and accurately evaluate the operating status of server components, and intuitively reflect their impact on the quality of server business processing in the form of initial component scores, thereby improving the accuracy of predicting server failures.

[0075] As an optional embodiment, predicting target fault information of the server according to the target service information of the plurality of server components includes:

[0076] Filtering a target component from the server components according to the target business information, wherein the target business information corresponding to the target component is used to indicate that the degree of influence of the current running state of the target component on the service handling quality of the server is greater than the target influence degree;

[0077] The target fault information corresponding to the target business information and the target component is screened out from the business information, component and fault information having corresponding relationships.

[0078] Optionally, in an embodiment of the present application, the target component can be the component with the lowest initial component score among the server components, or it can be the component with an initial component score lower than a set threshold among the server components, and then the corresponding target fault information can be queried based on the target business information and server component as indexes in a mapping table containing business information, components, and fault information.

[0079] Optionally, in an embodiment of the present application, after determining the target component, the business interaction information of the target component can be detected, wherein the business interaction information is used to indicate the business interaction relationship between the target component and other server components; the operation information of the associated component is queried according to the business interaction information, wherein the associated component is a component that has a business association relationship with the target component; and the target fault information is determined according to the operation information of the associated component. For example, after determining that the current target component is the CPU, the score of the component that has a business association relationship with the CPU, such as the memory, the cooling fan, etc., can be further queried, and then the operation information of the associated component, such as the read and write rate of the memory, the fan speed, etc., can be queried, and then the target fault information can be determined based on the operation information of the associated component.

[0080] Through the above method, it is possible to screen out target components that have a significant impact on the quality of business processing based on the target business information of multiple server components, and by querying the operating information of components that have business-related relationships with the target components, it is possible to more accurately find abnormally operating components in the server components, thereby improving the accuracy of predicting target fault information.

[0081] As an optional embodiment, after predicting the target fault information of the server according to the target service information of the plurality of server components, the method further includes:

[0082] Generate fault maintenance information for the server according to the target service information, wherein the fault maintenance information is used to prevent the server from having a fault indicated by the target fault information during the current service handling process;

[0083] Perform fault maintenance on the server according to the fault maintenance method indicated by the fault maintenance information.

[0084] Optionally, in an embodiment of the present application, corresponding fault maintenance information may be generated for the server based on the target fault information predicted by the target business information. For example, when the target fault information is determined to be CPU overheating, the maintenance information may include cleaning the radiator, replacing the cooling fan, adjusting the CPU load distribution, etc.

[0085] Optionally, in an embodiment of the present application, fault maintenance is performed on the server in accordance with the fault maintenance method indicated by the fault maintenance information. The fault maintenance information may be sent to the server operation and maintenance personnel, and the notification content may include the fault prediction results, maintenance suggestions, and the execution priority of the suggestions. The system may also automatically perform some simple or emergency maintenance measures based on the fault maintenance information. For example, when the CPU is overloaded, it may automatically trigger the adjustment of the CPU load and switch to backup hardware resources. When the CPU is overheated, it may automatically trigger the cooling system enhancement mode.

[0086] Through the above method, the potential failure of the server can be predicted based on the target business information, and specific fault maintenance information can be generated to guide the operation and maintenance personnel or the automated decision-making module to perform effective fault maintenance, thereby avoiding or reducing the impact of the failure on the current business processing process.

[0087] As an optional embodiment, generating fault maintenance information for the server according to the target service information includes:

[0088] Filtering a reference component from the server according to a target fault type to which the fault indicated by the target service information belongs, wherein the reference component is a server component associated with the operation fault indicated by the target fault information;

[0089] The fault maintenance information is generated according to the current operation information of the reference component.

[0090] Optionally, in an embodiment of the present application, after determining the fault type of the target component, reference components associated with the fault type can be screened out from the server. For example, if the target business information indicates a "CPU overheat" fault type, then the reference components may include a CPU, a radiator, a fan, a power supply unit, and a temperature sensor, etc., and then the current operating information of the reference components is further analyzed, which may include real-time operating data and historical operating data of the reference components. For example, when the CPU is overheated, the fan speed of the radiator, the voltage stability of the power supply unit, the reading of the temperature sensor, etc. can be analyzed. For historical data, the system can generate an operating information change curve, analyze the trend of the component operating status, and determine whether it is in a deterioration stage.

[0091] Optionally, in the embodiment of the present application, fault maintenance information may be generated based on the operating information of the reference component. For example, if the fan speed of the radiator is lower than the expected value, the fault maintenance information may be to check whether the fan is blocked or damaged; if the temperature reading displayed by the temperature sensor is inconsistent, the fault maintenance information may be to replace the sensor to ensure the accuracy of the reading.

[0092] Through the above method, it is possible to intelligently identify the fault type based on the target business information and the operation information of the server components, screen out the associated reference components, and generate specific fault maintenance information, thereby improving the maintenance efficiency of the server faults.

[0093] As an optional embodiment, the generating the fault maintenance information according to the current operation information of the reference component includes one of the following:

[0094] In the case where the target fault type is a processor occupancy fault, generating an occupancy adjustment operation for the processor according to the initial occupancy of each processor in the plurality of processors deployed in the server, wherein the reference component includes the plurality of processors, the operation information includes the initial occupancy, the fault maintenance information includes the occupancy adjustment operation, and the occupancy adjustment operation is used to adjust the difference between the occupancies of the plurality of processors by adjusting the occupancy of each processor;

[0095] In the case where the target fault type is server operation overtemperature, generating an operation parameter adjustment operation for the heat dissipation device deployed in the server according to an initial heat dissipation of the heat dissipation device, wherein the reference component includes the heat dissipation device, the operation information includes the initial heat dissipation, the fault maintenance information includes the operation parameter adjustment operation, and the operation parameter adjustment operation is used to adjust the heat dissipation of the heat dissipation device;

[0096] In the case where the target fault type is a target memory operation fault, the resource occupancy of a reference memory deployed in the server is obtained to generate a data transfer operation for the target memory, wherein the reference memory is a memory other than the target memory among multiple memories deployed in the server, the reference component includes the reference memory, the operation information includes the resource occupancy, the fault maintenance information includes the data transfer operation, and the data transfer operation is used to isolate the target memory and store the data stored in the target memory into the reference memory.

[0097] Optionally, in an embodiment of the present application, server failure conditions can be automatically responded to and processed. For example, when a failure condition occurs in which the occupancy rate of a certain processor is too high, such as the occupancy rate of a certain processor exceeds 80% for a long time, while the occupancy rates of other processors are below 50%, this indicates that there is a problem of uneven processor occupancy. At this time, the system will automatically calculate the difference between the occupancy rates of each processor to evaluate the balance of load distribution, and then migrate some tasks on the high-occupancy processor to the processor with lower load, or reduce the performance of the high-occupancy processor through throttling technology to balance the load between processors.

[0098] Optionally, in an embodiment of the present application, when an over-temperature fault occurs in the server, the system will automatically collect and analyze the initial heat dissipation of the cooling device, including parameters such as fan speed, radiator efficiency, and coolant flow rate, and then generate an operating parameter adjustment operation for the cooling device based on the initial heat dissipation analysis results. For example, if the fan speed is detected to be lower than expected, the operating parameter adjustment operation may be to increase the fan speed to enhance air flow, thereby increasing cooling efficiency.

[0099] Optionally, in an embodiment of the present application, when a certain storage device in a server fails in operation, the system will automatically select a reference storage device with lower resource usage from multiple storage devices deployed in the server, and then migrate the critical data being processed or stored in the target storage device to the reference storage device with lower resource usage, and isolate the faulty storage device from the current business process to prevent the fault from spreading or data loss.

[0100] Through the above method, by automatically generating corresponding fault maintenance information for different fault types, timely and effective measures are taken to prevent or reduce the impact of faults on server operation and business processing quality, thereby improving the efficiency of monitoring server faults.

[0101] In order to better understand the above process, the above process is described below in conjunction with an optional embodiment, but it is not intended to limit the technical solution of the embodiment of the present application.

[0102] In this embodiment, a method, apparatus, device and readable medium for evaluating the health score of a server are provided to score the health status of the entire machine. The user can be intuitively informed of the health status of the entire machine in the form of a score, which is more conducive to the customer's understanding and identification of the health of the entire machine. Figure 3 is a design diagram of a method for evaluating server health score according to an embodiment of the present application. The design method is as follows Figure 3 The shown modules include: health indicator collection module, health assessment algorithm module, health scoring system module, and automated decision-making module.

[0103] Among them, the health indicator collection module will collect various indicator data of the server, such as CPU usage, memory usage, disk space, network traffic, whole machine sensor, Pcie component status, PSU power supply status, etc.

[0104] The health assessment algorithm module is based on the indicator data collected by the health indicator collection module. The system will run a series of health assessment algorithms to determine the health status of the server. These algorithms involve threshold detection, trend analysis, anomaly detection, BIOS fault reporting and other technologies.

[0105] The health score system module will calculate a health score for the server component based on the result of the health assessment algorithm module, and then calculate the health score of the entire machine based on the defined component weight matrix and health score.

[0106] The automated decision module is based on the health score results of the server, and the BMC can perform automated processing, such as automatically triggering CPU throttling, adjusting load balancing strategies, sending alarms to notify users, etc., to maintain system stability and availability.

[0107] The present application provides a method, device, equipment and readable medium for evaluating the health score of a server. The health indicator collection module collects the health data of the whole machine components; the health assessment algorithm module scores different components; the health scoring system generates the final health score of the whole machine according to the matrix matching of different component scores; the automated decision module performs automated processing based on the health score of the whole machine to ensure the healthy operation of the whole machine. The method can systematically and accurately evaluate and predict the health of the whole machine, which is conducive to the user's physical examination to intervene in the health management of the server to improve the stability of the server.

[0108] Compared with the traditional health detection mechanism, integrating various server health indicators and calculating the server health score based on the health assessment algorithm can obtain an intuitive and reliable health score, making it easier for customers to obtain the health status of the server, such as Figure 3 As shown, the method specifically includes:

[0109] Step S31: The health indicator collection module collects various indicator data of the server, such as CPU usage, memory usage, disk space, network traffic, whole machine sensor, Pcie component status, PSU power supply status, etc. Figure 4 is a schematic diagram of a method for collecting health indicators according to an embodiment of the present application, such as Figure 4As shown in the figure, for the abnormal state of Intel CPU, BMC can directly access the internal Bank status register and CSR register of CPU through PECI to obtain it; for the abnormal state of memory, BMC can obtain it by accessing the SPD of memory through I3C; for the PCIE component state, BMC can obtain the health status according to the protocol supported by the specific component, for example, through MCTP, I2C, NVME-MI and other commands to obtain its health status register; for the health state of CPLD, BMC generally obtains it by accessing the status register through I2C; for the abnormal state of PSU, BMC constructs the health status word through PMbus to obtain its health status; for fans, the operation status of each fan rotor is generally detected through TACH; sensors are various temperature, humidity and pressure sensors integrated on the server motherboard, and BMC can generally access and obtain their information through I2C, and judge whether they are abnormal according to the defined threshold; for system information such as CPU utilization, memory utilization, disk space, network traffic, etc., an Agent is generally automatically run when the server is turned on, and the Agent transmits this part of system information to BMC through IPM B command.

[0110] Step S32: Figure 5 is a schematic diagram of a weight matrix scoring according to an embodiment of the present application, such as Figure 5 As shown in the figure, the health assessment algorithm module is based on the indicator data collected by the health indicator collection module. The system will run a series of health assessment algorithms to determine the health status of the server, and finally obtain the final score based on the weight matrix. For CPU, memory, and Pcie devices, BMC uses a combination of abnormal detection and BIOS fault reporting to obtain fault conditions and fault conditions, and then Figure 5 The corresponding weight value in the calculation of health score; for the status of sensors and fans, BMC uses threshold detection, trend analysis and other algorithms to calculate the status, and then calculates the health score based on the Figure 5 The health score is calculated based on the corresponding weight value in the matrix. For CPLD and PSU, the number of failures is generally calculated through statistical analysis, and then the health score is obtained by matching the weight matrix.

[0111] Step S33: Figure 6 is a workflow diagram of a health score arbitrator according to an embodiment of the present application, such as Figure 6 As shown in the figure, the health scoring system module obtains the score of each detection point based on the scoring results output by the health assessment algorithm module, and obtains the final health score of the server through the health scoring arbitrator. The whole process is as follows Figure 6 The health score arbitrator generally takes the maximum value from the scores corresponding to each detection point, which is the health score of the entire machine.

[0112] Step S34: The automated decision module is based on the health score results of the server. The BMC can perform automated processing. For example, when the CPU usage is too high, the CPU throttling is automatically triggered and the load balancing strategy is adjusted. When the sensor temperature is too high, the fan speed adjustment function is enabled. For the problem of high disk usage, an alarm can be sent to notify the user. For memory abnormalities, the faulty memory can be isolated by memory isolation. When the CPLD has an I ERR, the system can be restarted. The system is restored to health through the automated decision module and the health score is updated.

[0113] Figure 7 is a schematic diagram of a hardware link topology according to an embodiment of the present application, such as Figure 7 As shown, the BMC collects various indicator data of the server through the health indicator collection module, and then calculates the health score of each monitoring point through the health assessment algorithm module. The health scoring system module calculates the final health score of the server through the weight matrix and the health score arbitrator. Finally, the automated decision module executes different strategies based on the health score results of the server to optimize the health score and keep the whole machine stable. It can be seen that the method, device, equipment and readable medium for evaluating the health score of a server disclosed in the embodiment of the present application can accurately and timely evaluate the health score of the server, thereby providing users with risk control and improving the stability of the target server.

[0114] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0115] In this embodiment, a server fault prediction device is also provided, which is used to implement the above embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0116] Figure 8 is a structural block diagram of a server fault prediction device according to an embodiment of the present application, such as Figure 8 As shown, the device comprises:

[0117] An acquisition module 82 is used to acquire a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process;

[0118] A first prediction module 84 is used to predict target service information of each server component according to the operation information, wherein the target service information is used to indicate the influence of the current operation state of the server component on the quality of server service handling;

[0119] The second prediction module 86 is used to predict the target fault information of the server according to the target business information of the plurality of server components, wherein the target fault information is used to indicate the operation fault that will occur in the server during the current business processing.

[0120] Through the above device, by obtaining a variety of operating information generated by the server during the current business handling process, since the operating information can reflect the operating status of each component in the server during the current business handling process, it is possible to predict the target business information indicating the impact of each server component on the server business handling quality under the current operating state according to the acquired operating information of each server component, that is, the impact of the current operating state of each server component on the server business handling quality, and then, based on the predicted target business information of multiple server components, further predict the target fault information that characterizes the operating fault that will occur in the server during the current business handling process. Through the above method, the potential failure risk of the server can be predicted in advance according to the operating status of each server component in the server, and then corresponding prevention and treatment measures can be taken in advance according to the predicted fault information, so as to avoid the impact of passive perception on the server business after the server failure occurs. Therefore, the problem of low efficiency of server fault monitoring in the related technology can be solved, and the effect of improving the efficiency of server fault monitoring can be achieved.

[0121] Optionally, the first prediction module includes:

[0122] A first generating unit, configured to generate an initial component score of the server component according to the operation information, wherein the initial component score is used to indicate the influence of the operation state indicated by the operation information on the current service handling quality of the server;

[0123] A calculation unit is used to perform weighted calculation on the initial component score using the initial weight parameters corresponding to the server component to obtain a target component score of the server component, wherein the target business information includes the target component score, and the initial weight parameters are used to indicate the influence of the component function of the server component on the business handling quality of the server.

[0124] Optionally, the first generating unit is further used to:

[0125] In the case where the operation information is the first information, the first information is matched with a preset information range to obtain a target information range within which the first information falls, wherein the first information is the operation data reported by the server component when it is determined that the server component is currently in an abnormal operation state; and the component score configured corresponding to the target information range is determined as the initial component score;

[0126] In the case where the operation information is the second information, an operation information change curve of the server component before the current moment is generated according to the second information, wherein the second information is the operation data of the server component collected before the current moment, and the operation information change curve is used to indicate the change of the operation data of the server component in a continuous time period before the current moment; the operation information change curve is matched with a preset information change curve to obtain a target information change curve whose matching degree with the operation information change curve is greater than or equal to a target matching degree; and a component score configured corresponding to the target information change curve is determined as the initial component score;

[0127] In the case where the operation information is third information, the initial component score corresponding to the third information is determined from information and component scores having a corresponding relationship, wherein the third information is used to indicate the number of times the server component is in an abnormal operation state during the server handling of business.

[0128] Optionally, the second prediction module includes:

[0129] A first screening unit, configured to screen out a target component from the server component according to the target business information, wherein the target business information corresponding to the target component is used to indicate that the degree of influence of the current running state of the target component on the service handling quality of the server is greater than a target influence degree;

[0130] The second screening unit is used to screen out the target fault information corresponding to the target business information and the target component from the business information, component and fault information having a corresponding relationship.

[0131] Optionally, the device further comprises:

[0132] A generating module, configured to generate fault maintenance information for the server according to the target business information after predicting the target fault information of the server according to the target business information of the plurality of server components, wherein the fault maintenance information is used to prevent the server from having a fault indicated by the target fault information during a current business handling process;

[0133] A maintenance module is used to perform fault maintenance on the server according to the fault maintenance method indicated by the fault maintenance information.

[0134] Optionally, the generating module includes:

[0135] A third screening unit, configured to screen out reference components from the server according to a target fault type to which the fault indicated by the target service information belongs, wherein the reference component is a server component associated with the operation fault indicated by the target fault information;

[0136] The second generating unit is used to generate the fault maintenance information according to the current operation information of the reference component.

[0137] Optionally, the second generating unit is further used to:

[0138] In the case where the target fault type is a processor occupancy fault, generating an occupancy adjustment operation for the processor according to the initial occupancy of each processor in the plurality of processors deployed in the server, wherein the reference component includes the plurality of processors, the operation information includes the initial occupancy, the fault maintenance information includes the occupancy adjustment operation, and the occupancy adjustment operation is used to adjust the difference between the occupancies of the plurality of processors by adjusting the occupancy of each processor;

[0139] In the case where the target fault type is server operation overtemperature, generating an operation parameter adjustment operation for the heat dissipation device deployed in the server according to an initial heat dissipation of the heat dissipation device, wherein the reference component includes the heat dissipation device, the operation information includes the initial heat dissipation, the fault maintenance information includes the operation parameter adjustment operation, and the operation parameter adjustment operation is used to adjust the heat dissipation of the heat dissipation device;

[0140] In the case where the target fault type is a target memory operation fault, the resource occupancy of a reference memory deployed in the server is obtained to generate a data transfer operation for the target memory, wherein the reference memory is a memory other than the target memory among multiple memories deployed in the server, the reference component includes the reference memory, the operation information includes the resource occupancy, the fault maintenance information includes the data transfer operation, and the data transfer operation is used to isolate the target memory and store the data stored in the target memory into the reference memory.

[0141] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0142] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0143] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0144] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0145] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0146] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0147] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0148] The embodiments of the present application also provide a computer program, which includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any one of the above method embodiments.

[0149] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0150] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0151] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting server failures, characterized in that: include: Acquire a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process; Predicting target service information of each of the server components according to the operation information, wherein the target service information is used to indicate the influence of the current operation status of the server component on the quality of server service handling; The target fault information of the server is predicted based on the target business information of the plurality of server components, wherein the target fault information is used to indicate an operation fault that will occur in the server during the current business handling process.

2. The method according to claim 1, characterized in that The predicting target service information of each server component according to the operation information includes: Generating an initial component score of the server component according to the operation information, wherein the initial component score is used to indicate the impact of the operation state indicated by the operation information on the current service handling quality of the server; The initial component score is weightedly calculated using the initial weight parameters corresponding to the server component to obtain a target component score of the server component, wherein the target business information includes the target component score, and the initial weight parameters are used to indicate the influence of the component function of the server component on the business handling quality of the server.

3. The method according to claim 2, characterized in that Generating the initial component score of the server component according to the operation information includes one of the following: In the case where the operation information is the first information, the first information is matched with a preset information range to obtain a target information range within which the first information falls, wherein the first information is the operation data reported by the server component when it is determined that the server component is currently in an abnormal operation state; and the component score configured corresponding to the target information range is determined as the initial component score; In the case where the operation information is the second information, an operation information change curve of the server component before the current moment is generated according to the second information, wherein the second information is the operation data of the server component collected before the current moment, and the operation information change curve is used to indicate the change of the operation data of the server component in a continuous time period before the current moment; the operation information change curve is matched with the preset information change curve to obtain a target information change curve whose matching degree with the operation information change curve is greater than or equal to the target matching degree; and the component score configured corresponding to the target information change curve is determined as the initial component score; In the case where the operation information is third information, the initial component score corresponding to the third information is determined from information and component scores having a corresponding relationship, wherein the third information is used to indicate the number of times the server component is in an abnormal operation state during the server handling of business.

4. The method according to claim 1, characterized in that: The predicting target fault information of the server according to the target service information of the plurality of server components comprises: Filtering a target component from the server components according to the target business information, wherein the target business information corresponding to the target component is used to indicate that the degree of influence of the current running state of the target component on the service handling quality of the server is greater than the target influence degree; The target fault information corresponding to the target business information and the target component is screened out from the business information, component and fault information having corresponding relationships.

5. The method according to claim 1, characterized in that After predicting the target fault information of the server according to the target service information of the plurality of server components, the method further includes: Generate fault maintenance information for the server according to the target service information, wherein the fault maintenance information is used to prevent the server from having a fault indicated by the target fault information during the current service handling process; Perform fault maintenance on the server according to the fault maintenance method indicated by the fault maintenance information.

6. The method according to claim 5, characterized in that The generating fault maintenance information for the server according to the target service information includes: A reference component is screened out from the server according to a target fault type to which the fault indicated by the target business information belongs, wherein the reference component is a server component associated with the operation fault indicated by the target fault information; and the fault maintenance information is generated according to the current operation information of the reference component.

7. The method according to claim 6, characterized in that The generating the fault maintenance information according to the current operation information of the reference component includes one of the following: In the case where the target fault type is a processor occupancy fault, generating an occupancy adjustment operation for the processor according to the initial occupancy of each processor in the plurality of processors deployed in the server, wherein the reference component includes the plurality of processors, the operation information includes the initial occupancy, the fault maintenance information includes the occupancy adjustment operation, and the occupancy adjustment operation is used to adjust the difference between the occupancies of the plurality of processors by adjusting the occupancy of each processor; In the case where the target fault type is server operation overtemperature, generating an operation parameter adjustment operation for the heat dissipation device deployed in the server according to an initial heat dissipation of the heat dissipation device, wherein the reference component includes the heat dissipation device, the operation information includes the initial heat dissipation, the fault maintenance information includes the operation parameter adjustment operation, and the operation parameter adjustment operation is used to adjust the heat dissipation of the heat dissipation device; In the case where the target fault type is a target memory operation fault, the resource occupancy of a reference memory deployed in the server is obtained to generate a data transfer operation for the target memory, wherein the reference memory is a memory other than the target memory among multiple memories deployed in the server, the reference component includes the reference memory, the operation information includes the resource occupancy, the fault maintenance information includes the data transfer operation, and the data transfer operation is used to isolate the target memory and store the data stored in the target memory into the reference memory.

8. A server failure prediction device, characterized in that: include: An acquisition module, used to acquire a variety of operation information generated by the server during the current business processing process, wherein each type of operation information corresponds to a server component deployed in the server, and each type of operation information is used to indicate the operation status of the corresponding server component during the current business processing process; The first prediction module is used to predict the target business information of each of the server components based on the operation information, wherein the target business information is used to indicate the impact of the current operation status of the server component on the quality of server business processing; the second prediction module is used to predict the target fault information of the server based on the target business information of multiple server components, wherein the target fault information is used to indicate the operation fault that will occur in the server during the current business processing.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 7 are implemented.

Images